Detect Sip Flooding Attacks In Volte By Utilizing And Compressing Counting Bloom Filter

Mingli Wu,Na Ruan,Shiheng Ma,Haojin Zhu,Weijia Jia,Qingshui Xue,Songyang Wu
DOI: https://doi.org/10.1007/978-3-319-60033-8_12
2017-01-01
Abstract:As a new generation voice service, Voice over LTE (VoLTE) has attracted worldwide attentions in both the academia and industry. Different from the traditional voice call based on circuit-switched (CS), VoLTE evolves into the packet-switched (PS) field, which is quite open to the public. Though designed rigorously, similar to VoIP service, VoLTE also suffers from SIP (Session Initiation Protocal) flooding attacks. In this paper, two schemes inspired by Counting Bloom Filter (CBF) are proposed to thwart these attacks. In scheme I, we leverage CBF to accomplish flooding attack detection. In scheme II, we design a versatile CBF-like structure, PFilter, to achieve the same goal. Compared with previous relevant works, our detection schemes gain advantages in many aspects including low-rate flooding attack and stealthy flooding attack. Moreover, not only can our schemes detect the attacks with high accuracy, but also find out the attacker to ensure normal operation of VoLTE. Extensive experiments are performed to well evaluate the performance of the proposed two schemes.
What problem does this paper attempt to address?